(02-05-2013, 05:24 PM)Yorkie Wrote: A lot of my test gear over the years has been home-brew . . . . . . etc., etc. . . . . I guess I must have OCD or some such thing!
Then that makes two of us!
Yes, I do have a few items of test gear that come into the 'professional laboratory' category: most of these were acquired relatively cheaply by virtue of years of employment in various electronics firms: they had either been deemed B.E.R., redundant, of even salvaged from the skip! However, those few items aside, there is something rewarding & emotionally satisfying in meeting a 'test need' by designing and / or constructing an item that is home-brew: you can include only those features that you want (and omit those that you don't), and you can choose the construction such that future maintenance will not be a headache! For me, even if the finished article, in terms of cost, is comparable to a 'ready-made' commercial item, those considerations carry a lot of weight. Then there are those situations where what is wanted (to meet a need) is simply not available, and such items are occasionally quite simple affairs.
As for calibration and precision (these two are not the same, although they are often regarded as synonymous), of home-brew items, I take a similar attitude to David: will the finished article meet my needs in those two categories? Only today I have been tinkering around with a home-brew d.c. PSU 'add-on' in which it is necessary to monitor (and be able to adjust) the d.c. current output from it. Such is the intended purpose of this item that an indication of 300 mA ± 10 mA will be perfectly adequate. For this monitoring, I simply re-scaled an ancient moving-coil milliammeter (salvaged from the ubiquitous 'junk box') and added a suitable shunt. To calibrate it, I simply used two Fluke DMMs, an AVO 8 and a Phillips DVM, all connected in series with the meter. Those 4 items of test kit all read very nearly the same when 500 mA was flowing: the differences were insignificant. If the differences had been of a relative significance to the requirement, I would have taken the mean value of those separate indications. Hence, the resultant calibration of my 'monitoring meter' will be quite adequate for the need. When this item is finished, I may do a write-up about it and present same here: but, in the meantime, don't hold your breath!
Al.
